For rectal and head and neck cancers, direct pO2 measurements have established a significant intratumoral hypoxic fraction. This is of great clinical significance as tumor hypoxia is thought to negatively influence cancer response to chemoradiation (CMT), a major treatment component for these tumors. However, current methods for assessing tumor hypoxia are limited by their invasive nature and associated sampling errors due to heterogeneity of intratumoral oxygenation. The focus of this project is the clinical evaluation of non-invasive imaging of global tumor hypoxia via positron emission tomography (PET) in patients with rectal and head and neck cancers. We propose to evaluate two hypoxia tracers conjugated with positron-emitting radioisotopes: 18F-FMISO and 124I- IAZGP. Given their different characteristics, difficult-to-predict pharmacokinetics, and possible disease and anatomyrelated influences, we will test them in these two tumor types and identify the optimal tumor-specific tracer. Therefore, the Specific Aims of RP4 are: I. To identify the optimal hypoxia tracer for PET imaging of locally advanced rectal cancer and head and neck cancers. 100 patients of each disease site will be studied, Each patient will be imaged with 18F-FMISOand 124I-IAZGP. The images will be evaluated in terms of image quality and prognostic value of treatment outcome. II. To correlate, in rectal cancer, global tumor hypoxia assessed by PET imaging with that assessed by direct pO2 measurements using polarographic electrodes and IHC analysis of hypoxia-related proteins. We believe that these series of clinical studies will demonstrate the utility of PET hypoxia imaging and enhance the management of patients with rectal and head and neck cancers.